The utility of assessing the gross appearances of FNA specimens.

OBJECTIVE: Ideally, fine needle aspiration (FNA) cytology should be performed with near-patient assessment of the adequacy of the specimen by a cytopathologist. However, this is often not feasible. A cruder alternative is for the FNA practitioner to examine the gross appearances of the specimen and to try to predict the its quality. This study set out to determine the value of this approach. METHODS: The study was conducted in tertiary public hospitals in New Zealand and the UK. FNA gross material grading was performed by a variety of pathologists on FNA samples taken using manual guidance and image guidance. The FNA gross material grade was compared with the findings on microscopic examination. RESULTS: Nine out of 123 FNA samples were assessed as Grade 1 (unlikely to contain diagnostic material). All were subsequently reported as having insufficient diagnostic tissue on microscopic examination. Forty-two of the FNA samples were assessed as Grade 2 (possibly contains diagnostic material) and 46 as Grade 3 (probably contains diagnostic material). None from either of these grades was reported as showing insufficient diagnostic material on microscopic examination. Twenty-six cases were reported as Grade 4 (material suggesting a specific diagnosis). None of these was reported as showing insufficient diagnostic material on microscopic examination. The most common Grade 4 provisional diagnosis was that of a colloid cyst or colloid nodule of the thyroid (seven cases). Only two cases had misleading Grade 4 provisional diagnoses. Both were thought to be pus on gross examination but showed necrotic carcinoma on microscopic examination. CONCLUSIONS: The gross appearances of FNA samples can usually predict the adequacy of the samples and sometimes predict the final microscopic diagnosis. However, near-patient microscopic assessment of FNA specimens is preferable if available.

The gross appearances of fine needle aspiration cytology samples.

AIMS: This study set out to photograph and describe the gross appearances of fine needle aspiration (FNA) cytology samples of commonly encountered lesions. METHODS: During a 2 year period, a cytopathologist photographed the gross appearances of near patient FNA samples, concentrating on commonly encountered lesions. RESULTS: The gross appearances are described, accompanied by photographic illustrations. CONCLUSIONS: This paper describes and illustrates the gross appearances of FNA cytology samples of some commonly encountered lesions.

Immunoflow cytometry compared with PCR for the identification of clonality in FNAs of T-cell-rich B-cell lymphomas.

OBJECTIVE: The study aimed to compare the utility of immunoflow cytometry (IFC) with two IgH polymerase chain reaction (PCR) methods for the identification of clonality in fine needle aspirations (FNAs) from T-cell-rich B-cell lymphomas (TCRBCLs). METHODS: Ten cases of TCRBCLs were identified in which IFC had been performed according to our previously described method. Seven of these were cases in which the original diagnosis had been made by FNA cytology with IFC and cell block immunohistochemistry (IHC). The remaining three cases only had biopsies with histology, IFC and IHC. Formalin-fixed paraffin-embedded FNA cell block or histology tissue from these specimens had also been submitted for IgH PCR clonality studies using primers FR2a/VLJH and primers FR3a/VLJH. The results were reviewed and compared. RESULTS: All 10 case demonstrated B-cell clonality for at least one of the primer sets on PCR, but none showed light chain restriction on IFC. All TCRBCLs were positive for CD20 and CD79a but negative for CD10 and BCl-2. They were also consistently negative for CD22, CD23, CD5, CD43, ALK-1, cyclin D1 and CD30. CONCLUSIONS: If IgH PCR clonality assays had a turnaround time of 1 or 2 days, there might be a strong case for these studies supporting or even replacing IFC, in the FNA diagnosis of lymphoid lesions.

Fine needle aspiration cytology in the diagnosis of uncommon types of lymphoma.

AIMS: Fine needle aspiration (FNA) cytology is an accepted means of diagnosing and typing common forms of lymphoma, particularly small lymphocytic lymphoma, follicular lymphoma, and large B cell lymphoma. However, its usefulness for diagnosing less common forms of lymphoma is not clearly established and this study was designed to examine this. METHODS: The study reviewed the FNAs of suspected lymphomas collected over a period of approximately five years. RESULTS: FNA samples were available for 138 definite lymphomas; most were common forms of B cell lymphoma. However, there was also one Burkitt lymphoma (BL), two Burkitt-like large B cell lymphomas, 15 classic Hodgkin lymphomas (HLs), two nodular lymphocyte predominant Hodgkin lymphomas, four mantle cell lymphomas, two mediastinal (thymic) large B cell lymphomas (MLBCLs), 11 peripheral T cell lymphomas (PTCLs), and five T cell rich large B cell lymphomas (TCRLBCLs). CONCLUSIONS: FNA diagnosis of BL was possible with immunoflow cytometry (IFC), cell block immunohistochemistry (IHC), and cell block fluorescent in situ hybridisation for c-myc alteration. It was difficult to make a definite diagnosis of HL and MLBCL on FNA alone. Both tend to be sclerotic tumours and FNA tends to yield scanty neoplastic cells. The FNA diagnosis of PTCL depended on cell block IHC; IFC was not usually useful. TCRLBCL did not show light chain restriction on IFC of FNA samples, probably because of frequent reactive B cells in the tumour. Thus, HL, MLBCL, and TCRLBCL are often difficult to diagnose accurately on FNA cytology, even when using IFC and cell block IHC.

The detection of Simian virus 40 in mesotheliomas from New Zealand and England using real time FRET probe PCR protocols.

AIMS: To detect Simian virus 40 (SV40) DNA in mesotheliomas from New Zealand and from England using novel real time FRET probe polymerase chain reaction (PCR) protocols. METHODS: Twenty four mesotheliomas from New Zealand (Central North Island) and 32 mesotheliomas from England (Greater Manchester region) were examined. Two real time FRET probe PCR protocols were optimised and their analytical sensitivity compared using dilutions of SV40 DNA. A conventional SV40 large tumour antigen protocol with detection by probe hybridisation and chemiluminescent Southern blotting was also optimised. RESULTS: Both real time PCR protocols had the same analytical sensitivity, detecting down to 10(-6) pg of SV40 DNA for each reaction, approximately one SV40 copy. All of the 56 mesothelioma samples contained amplifiable beta globin DNA, but none contained amplifiable SV40 DNA with the conventional large T antigen PCR-Southern blotting protocol, or the two real time FRET probe PCR protocols. The positive and negative controls gave the expected results. There was no evidence of inhibition. CONCLUSIONS: There is abundant evidence in the literature for the presence of SV40 in mesotheliomas. However, this study found no evidence of SV40 in mesotheliomas from England and New Zealand. The extensive use of SV40 contaminated polio vaccine in New Zealand does not seem to have resulted in SV40 associated mesotheliomas.

Immunoflow cytometry and cell block immunohistochemistry in the FNA diagnosis of lymphoma: a review of 73 consecutive cases.

AIMS: To review the results of 73 consecutive fine needle aspirations (FNAs) that were collected by a pathologist and analysed by immunoflow cytometry. Material for a cell block was also collected from some of these lesions. METHODS: The setting was a large general hospital in rural New Zealand. The FNAs were performed by a pathologist, or a radiologist for image guided localisations. Material for immunoflow cytometry was collected into RPMI and, when required, material for a cell block was collected into formalin. RESULTS: Of the 73 samples collected by FNA nine were inadequate. Light chain restriction could be demonstrated in most FNA samples from B cell lymphomas (28 of 30 adequate samples). The exceptions were two cases of T cell rich B cell lymphoma. Artefactual light chain restriction was seen occasionally in T cell lymphomas, presumably as a result of autoantibodies binding to the cell surfaces. It was possible to subtype most (18 of 30 adequate samples) B cell lymphomas as chronic lymphocytic leukaemia (CLL), follicle centre cell lymphoma (FCCL), or mantle cell lymphoma. The CD4 to CD8 ratio was not usually restricted in T cell lymphomas and coexpression of CD4 and CD8 was not usually found. Loss of pan-T cell antigens was seen in some T cell lymphomas. Four of the six T cell lymphomas and three of the four non-lymphoid malignacies were diagnosed with the aid of cell block immunohistochemistry. Only one of the four cases of Hodgkin's lymphoma showed Reed-Sternberg cells in the FNA smears. CONCLUSIONS: It is not always possible to characterise lymphomas as fully with FNA and immunoflow cytometry as is possible with biopsy histology and a full battery of modern investigations. Nevertheless, in the setting of a large rural general hospital immunoflow cytometry on FNA samples is a highly effective method of diagnosing and typing B cell lymphomas. Immunoflow cytometry is of little use for T cell lymphomas or Hodgkin's lymphomas. We advocate the use of cell block immunohistochemistry in preference to immunoflow cytometry for cases in which the cytological appearance of the specimen is overtly malignant but the differential diagnosis includes non-lymphoid malignancy.

Improved fine needle aspiration (FNA) cytology results with a near patient diagnosis service for breast lesions.

This study is a review of the quality of FNA cytology results for breast lesions approximately 18 months before and 10 months after a change from a rapid diagnosis FNA service with consultant pathologist aspirators to a conventional FNA service with clinician aspirators of varied experience. The setting was symptomatic breast clinic in a large hospital in rural New Zealand acting as a tertiary referral centre for a population of 550,000. The results were collected retrospectively and prospectively. The quality of results for pathologist aspirators (total 810) and clinician aspirators (total 403) was compared using the definitions of the NHS Breast Screening Program Guidelines for Cytology Procedures and Reporting in Breast Cancer Screening. There were statistically significant differences in specificity (biopsy cases only) with 73% for pathologists and 49% for clinicians, specificity (full) with 74% and 56%, inadequate rate with 23% and 37%, and complete sensitivity with 76% and 67%. The use of pathologist aspirators allowed the specimens to be reported in a few minutes. Specimens taken by clinicians took at least 30 min to report. The financial aspects of the two approaches are discussed. When compared with clinician aspirators, pathologist aspirators obtained better quality results and these were reported more quickly.

Microsatellite abnormalities in plasma of patients with breast carcinoma: concordance with the primary tumour.

AIMS: To detect microsatellite abnormalities in the primary tumours and plasma of patients with breast carcinoma. METHODS: Plasma was obtained from 17 breast carcinoma patients before surgery. Corresponding tumour and benign lymph node (control) samples for each of the carcinoma patients were obtained from paraffin blocks. DNA was extracted from the plasma samples and the paraffin embedded tissue using previously described methods. RESULTS: The 17 primary tumours showed two examples of loss of heterozygosity and three examples of microsatellite instability; the 17 plasma samples showed three and one, respectively. Many of the longer microsatellites (over 200 base pairs) were difficult to amplify from plasma. The investigations suggested that this was because of the highly fragmented nature of plasma DNA. Only one example of loss of heterozygosity and one example of microsatellite instability showed a concordant pattern in both primary tumour and plasma. These were both in the same patient. CONCLUSIONS: DNA mutations concordant with those in the primary carcinomas can occasionally be detected in the plasma of patients with breast carcinoma. However, the frequency would have to be markedly improved before this could be of any diagnostic value.

Mutations of p53 gene and SV40 sequences in asbestos associated and non-asbestos-associated mesotheliomas.

AIM: To examine mesotheliomas for a possible relation between p53 immunostaining, p53 gene mutation, simian virus 40 (SV40), and asbestos exposure. METHODS: Paraffin sections from 11 mesotheliomas were used for p53 immunostaining and also to extract DNA. This was analysed for the presence of mutations in exons 5 to 8 of the p53 gene using a "cold" single strand conformational polymorphism method, together with sequencing. The DNA from the paraffin sections was also used to search for SV40 sequences. A 105 base pair segment at the 3' of the SV40 large T antigen (Tag) was targeted and any PCR amplification products were sequenced to confirm that they were of SV40 origin. EDAX electron microscopic differential mineral fibre counts were performed on dried lung tissue at a specialist referral centre. RESULTS: The fibre counts showed that seven of the mesotheliomas were associated with abnormally high asbestos exposure. Of these, two showed p53 immunostaining, none showed p53 gene mutation, and five showed SV40. Of the four other mesotheliomas, three showed p53 immunostaining, one showed a (silent) p53 mutation, and none showed SV40. The difference in frequency of SV40 detection was significant at the p < 0.05 level. CONCLUSIONS: Immunostaining for the p53 gene was relatively common but p53 mutations were rare in this series. SV40 virus sequence was detected in five of seven asbestos associated mesotheliomas but in none of the non-asbestos-associated mesotheliomas. This suggests there may be a synergistic interaction between asbestos and SV40 in human mesotheliomas. A study with a larger number of cases is needed to investigate these observations further.

Single-strand conformational polymorphism (SSCP)-detected p53 gene mutations are a less sensitive marker of malignancy in pleural fluids than p53 immunostaining.

p53 immunostaining has been advocated as a marker of malignancy in pleural biopsies and serous fluids. The object of this study was to compare the sensitivity and specificity of p53 immunostaining for the detection of malignant cells in pleural fluids with a technique designed to detect p53 gene mutations in exons 5, 6, 7 and 8 by SSCP and nucleotide sequencing. Five out of eight pleural fluids containing adenocarcinoma showed p53 immunostaining and two of these also showed polymorphisms on SSCP and a mutation on sequencing. None of the 10 benign pleural fluids showed immunostaining for p53 or polymorphisms on SSCP. We believe that the poor sensitivity of p53 gene mutation by SSCP is mainly due to DNA from the background reactive cells 'swamping' the mutant DNA. We do not advocate its use as a diagnostic aid.

Improved FNA cytology results with a near patient diagnosis service for non-breast lesions.

AIM: To review fine needle aspiration (FNA) cytology from sites other than the breast a year before and a year after the introduction of a near patient FNA diagnosis (NPFD) service in which the FNA were performed by a pathologist and reported within a few minutes. METHODS: The setting was a large hospital in rural New Zealand. The year before the introduction of the NPFD service was examined retrospectively, and the year after prospectively. The pattern of use and the quality of the results before and after starting the NPFD service were compared. RESULTS: Time taken to report the specimens decreased from a few days to a few minutes. There were statistically significant changes in the following: an increase from 237 to 304 in the number of non-breast FNA performed, and in particular an increase from 65 to 113 in the number for general surgery; an increase in the use of immunolabelled flow cytometry from 0 to 19 and cell blocks from 3 to 41; an increase in specificity from 53% to 80%; a decrease in the overall inadequacy rate from 29% to 9%; and a decrease in the inadequacy rate for cancers from 9% to 2%. The cost of the non-breast FNA service increased by about 9200 Pounds a year. CONCLUSIONS: Starting an NPFD service for sites other than the breast greatly reduced the reporting time and produced statistically significant increases in the use of FNA cytology and in the quality of the results.

Mutations of p53 gene can be detected in the plasma of patients with large bowel carcinoma.

AIMS: To attempt to detect p53 gene mutations in the plasma of patients with large bowel carcinoma. METHODS: Plasma was collected from 20 control patients with no history of cancer and from 17 patients with large bowel carcinoma. Corresponding tumour and benign lymph node (control) samples for each of the carcinoma patients were obtained from paraffin blocks. A Dukes' stage was determined for each tumour. DNA was extracted from the plasma samples and the paraffin embedded tissue using previously described methods. A nested primer polymerase chain reaction protocol was used for the amplification of exons 5 to 8 of the p53 gene. "Cold" single strand conformational polymorphism (SSCP) was performed on mini gels and silver stained. Abnormal bands were excised, the DNA eluted, and reamplified for automated dye termination sequencing. Any sample showing an apparent mutation was rechecked from the original extracted DNA sample at least three times. RESULTS: p53 gene mutations were not found in the control specimens. They were found in both the primary tumour and the plasma in three cases, in the primary tumour alone in one case, and in the plasma alone in two cases. One of the latter two cases also had metastatic transitional cell carcinoma of the bladder and the other had widespread metastatic deposits. One of the cases with mutant DNA in both the plasma and the primary was a Dukes' stage B tumour. The others were Dukes' C and Dukes' D. CONCLUSIONS: p53 gene mutations can be detected in the plasma of some patients with large bowel carcinoma and these are concordant with those in the primary carcinomas.

p53 immunostaining is a highly specific and moderately sensitive marker of malignancy in serous fluid cytology.

Several studies have suggested that p53 immunostaining does not occur in benign mesothelium but is common in malignancies involving the serous surfaces, including malignant mesothelioma. As a result, p53 has been advocated as a marker of malignancy in serous fluid cytology. However, the specificity of p53 in this context has been brought into question by some studies that claim to have found immunostaining in benign mesothelium. The aim of our study was to examine p53 immunostaining in a large series of serous fluids to try to resolve the uncertainty. Monoclonal Do-7 antibody was used to immunostain ethanol-fixed cytospin preparations employing an alkaline phosphatase method. Positivity was found in 17 of 35 malignant effusions, including two probable mesotheliomas, but was not found in any of 115 benign effusions. Our study suggests that p53 immunostaining is a highly specific and moderately sensitive marker of malignancy in serous fluids.

A review of 50 consecutive cytology cell block preparations in a large general hospital.

AIMS: To review consecutive cell block preparations of cytological specimens in a large general hospital. METHODS: 50 cell blocks were made over an 18 month period in which about 1900 fine needle aspirations (FNAs) were performed. The aspirator was a cytologist or, for image guided FNAs, a radiologist with a cytologist at hand to collect the specimen. Forty eight cell blocks were from FNAs and two were from serous fluids. RESULTS: The cellularity of the cell blocks was inadequate in only four preparations. The main motive for making cell blocks was to obtain tissue for immunohistochemistry. This was performed on 28 cases and a total of 107 immunostained sections were produced. The most common diagnostic dilemma was between carcinoma and melanoma, and the second most common between carcinoma and lymphoma. Consequently cytokeratin, S-100, and LCA were the most frequently used antibodies. At least one of these three antibodies was positive in 17 cases. Five cases were immunostained only for prognostic breast markers. CONCLUSIONS: The use of cell block immunohistochemistry is a reliable and technically unsophisticated aid in the cytological examination of tumours other than lymphomas. Success depends on having highly experienced aspirators that reliably obtain sufficiently cellular material.

A silver tattoo of the nasal mucosa after silver nitrate cautery.

We report a silver tattoo of the nasal mucosa that occurred after silver nitrate cautery for nasal bleeding. This type of tattoo is a very rare potential mimic of melanoma and appears not to have been described before. It has similar features to an amalgam tattoo of the oral mucosa on histology and energy dispersive X-ray analysis (EDAX).

Leiomyosarcoma of the tongue: a very rare tumour.

We describe a leiomyosarcoma of the tongue in a 60-year-old man. The diagnosis was supported by immunohistochemical positivity for desmin and alpha-1 smooth muscle actin.

p53 immunostaining suggests that uterine carcinosarcomas are monoclonal.

The histogenesis of carcinosarcomas has intrigued pathologists for a long time and remains unresolved. Two main theories have been put forward, one suggesting that they are monoclonal, another suggesting that they are biclonal. Our study examined p53 immunostaining in 17 uterine carcinosarcomas (mixed Müllerian tumours) and found positivity in five (30%). There was no disparity in immunostaining between the epithelial and the stromal components in any of the 17 tumours. This concordance in every tumour would be very unlikely if carcinosarcomas are biclonal. However, it would be expected if carcinosarcomas are monoclonal.